June 2012 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

APIC timer not working
by Vaibhav Jain June 25, 2012

June 25, 2012

Hi, I am trying to write my own kernel but I am facing problems with timer interrupt. I have been using PIC and PIT to generate timer interrupts and it has been working fine but now I want to switch to local APIC timer. I am using bochs as emulator. I have made some changes but all in vain. Here is what I have done : - compiled bochs with cpu_leve 6 to get the APIC enabled. - I have checked that APIC is present and is enabled (code at the end) - I have added code for disabling PIC. Now my code first remaps IRQs in PIC and then disables the PIC. If I don't remap the IRQs I sometimes get a double fault. - Added a page table entry for the APIC base - 0xFEE00000 which has the Cache Disable bit set. I haven't mapped it to a physical page. I have checked that the apic base address is correct by reading the APIC BASE MSR. (code at the end) - Added code from the wiki - http://wiki.osdev.org/APIC_timer to initialize APIC timer but it did not work.The timer interrupt didn't get triggered. - Now I have reduced the code from the wiki to only a few lines (code below) but the interrupt is just not getting generated. I checked bit 32 of the Interrupt Request register (0xFEE00390) and found it to be Set. But the interrupt handler is not getting called. The IDT and all the interrupt handlers are in place. If I switch back to the old code using PIC and PIT it works fine. Please let me know if I am doing something wrong. I am unable to figure out anything. *Code for APIC timer :* uint32 apic = 0xFEE00000; uint32 APIC_SPURIOUS = 0x0F0; uint32 APIC_SW_ENABLE = 0x100; uint32 APIC_DFR = 0x0E0; uint32 APIC_TASKPRIOR = 0x80; uint32 APIC_TMRDIV = 0x3E0; uint32 APIC_LVT_TMR = 0x320; uint32 TMR_PERIODIC = 0x20000; uint32 APIC_TMRINITCNT = 0x380; void apic_timer_init(){ * //software enable apic timer * *(uint32*)(apic+APIC_SPURIOUS)=0xFF|APIC_SW_ENABLE; * //set DFR to flat mode * *(uint32*)(apic+APIC_DFR)=0xFFFFFFFF; * //set task priority to 0 to allow all * *(uint32*)(apic+APIC_TASKPRIOR) = 0; * //set divider to 16 * *(uint32*)(apic+APIC_TMRDIV)=0x03; * //map APIC timer to an interrupt enable it in periodic mode * *(uint32*)(apic+APIC_LVT_TMR)=32|TMR_PERIODIC; * //initialize intial count * *(uint32*)(apic+APIC_TMRINITCNT)=0x10000; } *Code for checking if APIC is present:* ** //check if APIC is present using CPUID uint32 cpuHasAPIC() { uint32 eax , edx; eax = 0x1; uint32 CPU_FLAG_APIC = 0x200; asm volatile("cpuid": "=d"(edx) :"a"(eax)); return edx & CPU_FLAG_APIC; } ** ** *Code for checking if APIC is enabled and for getting APIC Base address :* ** void checkAPICStatus() { uint32 APIC_ENABLE_FLAG = 0x800; uint32 APIC_BASE_FLAG = 0xFFFFF000; uint32 ecx = 0x1B; * //<<<<<<<<<<<<<< APIC BASE MSR * uint32 eax, edx; asm volatile("rdmsr":"=a"(eax), "=d"(edx):"c"(ecx)); uint32 apic_enabled = eax & APIC_ENABLE_FLAG; if(apic_enabled) { print_str("\nAPIC IS ENABLED!!"); }else { print_str("\nAPIC IS DISABLED"); } *// Get APIC BASE address * uint32 apic_base = eax & APIC_BASE_FLAG; * // print APIC Base address* print_str("\n APIC BASE ADDRESS IS "); print_hex(apic_base); } - Thanks Vaibhav Jain

1 0

could some current.h files be simplified?
by Robert P. J. Day June 23, 2012

June 23, 2012

it *seems* as if a number of current.h files from various architectures could be simplified. here's <asm-generic/current.h>, which gives any architecture a generic starting point in defining both get_current() and current: #ifndef __ASM_GENERIC_CURRENT_H #define __ASM_GENERIC_CURRENT_H #include <linux/thread_info.h> #define get_current() (current_thread_info()->task) #define current get_current() #endif /* __ASM_GENERIC_CURRENT_H */ and if that's acceptable, any architecture is welcome to simply include it as, for instance, mips does in <asm/current.h>: #include <asm-generic/current.h> but here's the current.h file for parisc: #ifndef _PARISC_CURRENT_H #define _PARISC_CURRENT_H #include <linux/thread_info.h> struct task_struct; static inline struct task_struct * get_current(void) { return current_thread_info()->task; } #define current get_current() #endif /* !(_PARISC_CURRENT_H) */ i'm not sure i see why the generic version wasn't adequate for parisc (apart from the explicit pointer casting). same thing with the cris architecture and, i'm sure, others. is there any reason why some of those current.h files can't just include the generic one? rday -- ======================================================================== Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================

2 1

toolchain compilation error in gcc-3.4.4-2.16.1
by sanjay das June 23, 2012

June 23, 2012

Hello everyone, I was trying to compile toochain gcc-3.4.4-2.16.1 for Atheros AP83 board(toolchain have support for this) but after a certain point of time i get the below error where i am stuck .Please me know if anyone knows how to get it resolved and what is the issue behind this. ================================================================================================================ make[3]: Entering directory `/home/sanjay/MYBUILD/LSDK7.3.0.435/build/gcc-3.4.4-2.16.1/toolchain_build_mips_nofpu/gcc-3.4.4-initial/gcc' for d in libgcc; do \ if [ -d $d ]; then true; else /bin/sh /home/sanjay/MYBUILD/LSDK7.3.0.435/build/gcc-3.4.4-2.16.1/toolchain_build_mips_nofpu/gcc-3.4.4/gcc/mkinstalldirs $d; fi; \ done if [ -f stmp-dirs ]; then true; else touch stmp-dirs; fi rm -rf ./libgcc.a mips-linux-uclibc-ar rc ./libgcc.a libgcc/./_muldi3.o libgcc/./_negdi2.o libgcc/./_lshrdi3.o libgcc/./_ashldi3.o libgcc/./_ashrdi3.o libgcc/./_cmpdi2.o libgcc/./_ucmpdi2.o libgcc/./_floatdidf.o libgcc/./_floatdisf.o libgcc/./_fixunsdfsi.o libgcc/./_fixunssfsi.o libgcc/./_fixunsdfdi.o libgcc/./_fixdfdi.o libgcc/./_fixunssfdi.o libgcc/./_fixsfdi.o libgcc/./_fixxfdi.o libgcc/./_fixunsxfdi.o libgcc/./_floatdixf.o libgcc/./_fixunsxfsi.o libgcc/./_fixtfdi.o libgcc/./_fixunstfdi.o libgcc/./_floatditf.o libgcc/./_clear_cache.o libgcc/./_enable_execute_stack.o libgcc/./_trampoline.o libgcc/./__main.o libgcc/./_absvsi2.o libgcc/./_absvdi2.o libgcc/./_addvsi3.o libgcc/./_addvdi3.o libgcc/./_subvsi3.o libgcc/./_subvdi3.o libgcc/./_mulvsi3.o libgcc/./_mulvdi3.o libgcc/./_negvsi2.o libgcc/./_negvdi2.o libgcc/./_ctors.o libgcc/./_ffssi2.o libgcc/./_ffsdi2.o libgcc/./_clz.o libgcc/./_clzsi2.o libgcc/./_clzdi2.o libgcc/./_ctzsi2.o libgcc/./_ctzdi2.o libgcc/./_popcount_tab.o libgcc/./_popcountsi2.o libgcc/./_popcountdi2.o libgcc/./_paritysi2.o libgcc/./_paritydi2.o libgcc/./_divdi3.o libgcc/./_moddi3.o libgcc/./_udivdi3.o libgcc/./_umoddi3.o libgcc/./_udiv_w_sdiv.o libgcc/./_udivmoddi4.o libgcc/./_pack_sf.o libgcc/./_unpack_sf.o libgcc/./_addsub_sf.o libgcc/./_mul_sf.o libgcc/./_div_sf.o libgcc/./_fpcmp_parts_sf.o libgcc/./_compare_sf.o libgcc/./_eq_sf.o libgcc/./_ne_sf.o libgcc/./_gt_sf.o libgcc/./_ge_sf.o libgcc/./_lt_sf.o libgcc/./_le_sf.o libgcc/./_unord_sf.o libgcc/./_si_to_sf.o libgcc/./_sf_to_si.o libgcc/./_negate_sf.o libgcc/./_make_sf.o libgcc/./_sf_to_df.o libgcc/./_sf_to_tf.o libgcc/./_thenan_sf.o libgcc/./_sf_to_usi.o libgcc/./_usi_to_sf.o libgcc/./_pack_df.o libgcc/./_unpack_df.o libgcc/./_addsub_df.o libgcc/./_mul_df.o libgcc/./_div_df.o libgcc/./_fpcmp_parts_df.o libgcc/./_compare_df.o libgcc/./_eq_df.o libgcc/./_ne_df.o libgcc/./_gt_df.o libgcc/./_ge_df.o libgcc/./_lt_df.o libgcc/./_le_df.o libgcc/./_unord_df.o libgcc/./_si_to_df.o libgcc/./_df_to_si.o libgcc/./_negate_df.o libgcc/./_make_df.o libgcc/./_df_to_sf.o libgcc/./_df_to_tf.o libgcc/./_thenan_df.o libgcc/./_df_to_usi.o libgcc/./_usi_to_df.o libgcc/./unwind-dw2.o libgcc/./unwind-dw2-fde.o libgcc/./unwind-sjlj.o libgcc/./gthr-gnat.o libgcc/./unwind-c.o libgcc/./_eprintf.o libgcc/./__gcc_bcmp.o *** buffer overflow detected ***: mips-linux-uclibc-ar terminated ======= Backtrace: ========= /lib/i386-linux-gnu/libc.so.6(__fortify_fail+0x45)[0x4011c8d5] /lib/i386-linux-gnu/libc.so.6(+0xe66d7)[0x4011b6d7] /lib/i386-linux-gnu/libc.so.6(+0xe5d35)[0x4011ad35] /lib/i386-linux-gnu/libc.so.6(_IO_default_xsputn+0x91)[0x400a0f91] /lib/i386-linux-gnu/libc.so.6(_IO_padn+0xc9)[0x40094c89] /lib/i386-linux-gnu/libc.so.6(_IO_vfprintf+0x38e4)[0x40078a14] /lib/i386-linux-gnu/libc.so.6(__vsprintf_chk+0xc9)[0x4011ae09] /lib/i386-linux-gnu/libc.so.6(__sprintf_chk+0x2f)[0x4011ad1f] mips-linux-uclibc-ar[0x804e928] mips-linux-uclibc-ar[0x8050c94] mips-linux-uclibc-ar[0x8058f82] mips-linux-uclibc-ar[0x804c3ab] mips-linux-uclibc-ar[0x804a277] /lib/i386-linux-gnu/libc.so.6(__libc_start_main+0xf3)[0x4004e113] mips-linux-uclibc-ar[0x804a865] ======= Memory map: ======== 08048000-080e4000 r-xp 00000000 08:01 4243487 /export/tools-ap83fus/bin/mips-linux-uclibc-ar 080e4000-080e5000 r--p 0009b000 08:01 4243487 /export/tools-ap83fus/bin/mips-linux-uclibc-ar 080e5000-080e6000 rw-p 0009c000 08:01 4243487 /export/tools-ap83fus/bin/mips-linux-uclibc-ar 080e6000-080ea000 rw-p 00000000 00:00 0 09531000-09618000 rw-p 00000000 00:00 0 [heap] 40000000-4001e000 r-xp 00000000 08:01 1794064 /lib/i386-linux-gnu/ ld-2.13.so 4001e000-4001f000 r--p 0001d000 08:01 1794064 /lib/i386-linux-gnu/ ld-2.13.so 4001f000-40020000 rw-p 0001e000 08:01 1794064 /lib/i386-linux-gnu/ ld-2.13.so 40020000-40021000 r-xp 00000000 00:00 0 [vdso] 40021000-40023000 rw-p 00000000 00:00 0 40023000-4002a000 r--s 00000000 08:01 3662914 /usr/lib/i386-linux-gnu/gconv/gconv-modules.cache 40035000-401ab000 r-xp 00000000 08:01 1794067 /lib/i386-linux-gnu/ libc-2.13.so 401ab000-401ad000 r--p 00176000 08:01 1794067 /lib/i386-linux-gnu/ libc-2.13.so 401ad000-401ae000 rw-p 00178000 08:01 1794067 /lib/i386-linux-gnu/ libc-2.13.so 401ae000-401b2000 rw-p 00000000 00:00 0 401b2000-403b2000 r--p 00000000 08:01 3665681 /usr/lib/locale/locale-archive 403c4000-403e0000 r-xp 00000000 08:01 1794092 /lib/i386-linux-gnu/libgcc_s.so.1 403e0000-403e1000 r--p 0001b000 08:01 1794092 /lib/i386-linux-gnu/libgcc_s.so.1 403e1000-403e2000 rw-p 0001c000 08:01 1794092 /lib/i386-linux-gnu/libgcc_s.so.1 bfac5000-bfae8000 rw-p 00000000 00:00 0 [stack] make[3]: *** [libgcc.a] Aborted make[3]: *** Deleting file `libgcc.a' make[3]: Leaving directory `/home/sanjay/MYBUILD/LSDK7.3.0.435/build/gcc-3.4.4-2.16.1/toolchain_build_mips_nofpu/gcc-3.4.4-initial/gcc' make[2]: *** [libgcc.a] Error 2 make[2]: Leaving directory `/home/sanjay/MYBUILD/LSDK7.3.0.435/build/gcc-3.4.4-2.16.1/toolchain_build_mips_nofpu/gcc-3.4.4-initial/gcc' make[1]: *** [all-gcc] Error 2 make[1]: Leaving directory `/home/sanjay/MYBUILD/LSDK7.3.0.435/build/gcc-3.4.4-2.16.1/toolchain_build_mips_nofpu/gcc-3.4.4-initial' make: *** [/home/sanjay/MYBUILD/LSDK7.3.0.435/build/gcc-3.4.4-2.16.1/toolchain_build_mips_nofpu/gcc-3.4.4-initial/.compiled] Error 2 ====================================================================================================== Regards Sanjay Kumar Das

2 2

Kernel Memory
by Vijay Chauhan June 22, 2012

June 22, 2012

Hello, I am newbie. It has been said "kernel memory is not pageable" What does it mean? There is no concept of kernel virtual address? Any simple explanation will help me to udnerstand. Thanks, Vijay

6 6

explanation of __va() and __pa() macros.
by AFT June 22, 2012

June 22, 2012

Hi, My source is from 2.6 git tree. In LDD3 its said that a logical address is mapped to physical address and vice versa by the macros __va() and __pa(). In my source tree they are defined as follows: <include/asm-generic/page.h> #define __va(x) ((void *)((unsigned long) (x))) #define __pa(x) ((unsigned long) (x)) I'm not actually understanding how these "cute" looking casting is performing "address mapping". I thought address mapping involves more complex operations. cheers. aft

3 4

Re: Kernelnewbies Digest, Vol 19, Issue 43
by zhaoqiao June 22, 2012

June 22, 2012

2012/6/22 <kernelnewbies-request(a)kernelnewbies.org> > Send Kernelnewbies mailing list submissions to > kernelnewbies(a)kernelnewbies.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies > or, via email, send a message with subject or body 'help' to > kernelnewbies-request(a)kernelnewbies.org > > You can reach the person managing the list at > kernelnewbies-owner(a)kernelnewbies.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Kernelnewbies digest..." > > > Today's Topics: > > 1. Re: download the source code of some commands (Greg Freemyer) > 2. could some current.h files be simplified? (Robert P. J. Day) > 3. Re: download the source code of some commands (Wang Lei) > 4. Re: download the source code of some commands (??) > 5. Re: Kernel Memory (michi1(a)michaelblizek.twilightparadox.com) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 21 Jun 2012 10:42:12 -0400 > From: Greg Freemyer <greg.freemyer(a)gmail.com> > Subject: Re: download the source code of some commands > To: ?? <wangzhe5004(a)gmail.com> > Cc: kernelnewbies <kernelnewbies(a)kernelnewbies.org> > Message-ID: > <CAGpXXZLM7s_%3D2xWcEDFtSrj_0-EjgP7aP7XNu1XEmZMEhXT53w(a)mail.gmail.com > > > Content-Type: text/plain; charset=UTF-8 > > On Thu, Jun 21, 2012 at 8:48 AM, ?? <wangzhe5004(a)gmail.com> wrote: > > ? Hi all: > > ? ? ?? I want to see some source code of some commands,for > > example,halt,reboot,uptime,and so on. > > > > but i don't kown where to download??? can you give me some advice? > > > > Thanks in advance! > > You need to figure out how your distro of choice lets you query the > filelist for packages. > > I use openSUSE primarily and it has the Yast tool that lets you do > that in the software management section. > > uptime => coreutils > reboot => aaa_base > halt => aaa_base > > Once you know the package your distro uses you can download the > corresponding source package for it. > > You will have better luck learning the process you need to know by > asking on a distro specific mailing list. > > Greg > > hi, you can use the download tool to get the sources code. eg: apt-get source ssh (ubuntu) you can get ssh source code. but, not all ! > > > ------------------------------ > > Message: 2 > Date: Thu, 21 Jun 2012 10:54:59 -0400 (EDT) > From: "Robert P. J. Day" <rpjday(a)crashcourse.ca> > Subject: could some current.h files be simplified? > To: Kernel Newbies <kernelnewbies(a)kernelnewbies.org> > Message-ID: <alpine.DEB.2.02.1206211048240.22454@oneiric> > Content-Type: TEXT/PLAIN; charset=US-ASCII > > > it *seems* as if a number of current.h files from various > architectures could be simplified. here's <asm-generic/current.h>, > which gives any architecture a generic starting point in defining both > get_current() and current: > > #ifndef __ASM_GENERIC_CURRENT_H > #define __ASM_GENERIC_CURRENT_H > > #include <linux/thread_info.h> > > #define get_current() (current_thread_info()->task) > #define current get_current() > > #endif /* __ASM_GENERIC_CURRENT_H */ > > and if that's acceptable, any architecture is welcome to simply > include it as, for instance, mips does in <asm/current.h>: > > #include <asm-generic/current.h> > > but here's the current.h file for parisc: > > #ifndef _PARISC_CURRENT_H > #define _PARISC_CURRENT_H > > #include <linux/thread_info.h> > > struct task_struct; > > static inline struct task_struct * get_current(void) > { > return current_thread_info()->task; > } > > #define current get_current() > > #endif /* !(_PARISC_CURRENT_H) */ > > i'm not sure i see why the generic version wasn't adequate for > parisc (apart from the explicit pointer casting). same thing with the > cris architecture and, i'm sure, others. > > is there any reason why some of those current.h files can't just > include the generic one? > > rday > > -- > > ======================================================================== > Robert P. J. Day Ottawa, Ontario, CANADA > http://crashcourse.ca > > Twitter: http://twitter.com/rpjday > LinkedIn: http://ca.linkedin.com/in/rpjday > ======================================================================== > > > > ------------------------------ > > Message: 3 > Date: Thu, 21 Jun 2012 23:22:49 +0800 > From: Wang Lei <f3d27b(a)gmail.com> > Subject: Re: download the source code of some commands > To: ?? <wangzhe5004(a)gmail.com> > Cc: kernelnewbies <kernelnewbies(a)kernelnewbies.org> > Message-ID: <4fe33c55.83da440a.4026.ffffc75b(a)mx.google.com> > Content-Type: text/plain; charset=utf-8 > > If you use Debian or other distribution with apt package system. > > You can use whereis to locate this command. > [~]$ whereis halt > halt: /sbin/halt /usr/share/man/man5/halt.5.gz > /usr/share/man/man8/halt.8.gz > > Then, use dpkg -S to show which package it's included. This one is: > sysvinit. > [~]$ dpkg -S /sbin/halt > sysvinit: /sbin/halt > > Now, you can download the source from Debian source, use: > [~]$ apt-get source sysvinit > > Or if you want the original source, use aptitude (or apt-cache) find > it's homepage. > [~]$ aptitude show sysvinit > Package: sysvinit > ...... > Homepage: http://savannah.nongnu.org/projects/sysvinit > > At last, get it from there. As for other package systems, I think there > should be a way too. Or you can google. > > On 2012-06-21 20:48:48 +0800, ?? wrote: > > Hi all: > > I want to see some source code of some commands,for > > example,halt,reboot,uptime,and so on. > > > > but i don't kown where to download? can you give me some advice? > > > > Thanks in advance! > > _______________________________________________ > > Kernelnewbies mailing list > > Kernelnewbies(a)kernelnewbies.org > > http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies > > -- > Regards, > Lei > > > > ------------------------------ > > Message: 4 > Date: Thu, 21 Jun 2012 23:46:21 +0800 > From: ?? <wangzhe5004(a)gmail.com> > Subject: Re: download the source code of some commands > To: Wang Lei <f3d27b(a)gmail.com> > Cc: kernelnewbies <kernelnewbies(a)kernelnewbies.org> > Message-ID: > <CAJrRU3XDpS620jOe%2BQ8W%2BsiuGvR3CzzJOHZg3e%2BpWSSU5t0Tuw(a)mail.gmail.com > > > Content-Type: text/plain; charset="gb2312" > > 2012/6/21 Wang Lei <f3d27b(a)gmail.com> > > > If you use Debian or other distribution with apt package system. > > > > You can use whereis to locate this command. > > [~]$ whereis halt > > halt: /sbin/halt /usr/share/man/man5/halt.5.gz > > /usr/share/man/man8/halt.8.gz > > > > Then, use dpkg -S to show which package it's included. This one is: > > sysvinit. > > [~]$ dpkg -S /sbin/halt > > sysvinit: /sbin/halt > > > > Now, you can download the source from Debian source, use: > > [~]$ apt-get source sysvinit > > > > Or if you want the original source, use aptitude (or apt-cache) find > > it's homepage. > > [~]$ aptitude show sysvinit > > Package: sysvinit > > ...... > > Homepage: http://savannah.nongnu.org/projects/sysvinit > > > > At last, get it from there. As for other package systems, I think there > > should be a way too. Or you can google. > > > > Thank you very much ! i follow your steps, and download the source code > successfully. > > On 2012-06-21 20:48:48 +0800, ?? wrote: > > Hi all: > > I want to see some source code of some commands,for > > example,halt,reboot,uptime,and so on. > > > > but i don't kown where to download? can you give me some advice? > > > > Thanks in advance! > > > > _______________________________________________ > > > Kernelnewbies mailing list > > > Kernelnewbies(a)kernelnewbies.org > > > http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies > > > > -- > > Regards, > > Lei > > >

1 0

Sysfs class attribute problem
by jeshwanth Kumar N K June 21, 2012

June 21, 2012

Hello all I am new to sysfs interface ans I read about in mochel's documentation. And doing some experiements on it. Let's come to Tue problem, in my module I have a global variable type char* myglobal and it s static, I am putting that value in my one and only class attribute using show function. But for the same if I a user put some value in to the using echo or something, the attribute value is not changing. And one more thing s after I passed the value to the attribute my system not shutting down :( . Please help me how to read values. // in store function I am doing this: Myglobal = bus; Sent from my HTC

3 4

download the source code of some commands
by 王哲 June 21, 2012

June 21, 2012

Hi all: I want to see some source code of some commands,for example,halt,reboot,uptime,and so on. but i don't kown where to download? can you give me some advice? Thanks in advance!

6 8

RE: download the source code of some commands
by jeshwanth Kumar N K June 21, 2012

June 21, 2012

Hello U can download coreutils from gnu.org. Sent from my HTC -- Send Kernelnewbies mailing list submissions to kernelnewbies(a)kernelnewbies.org To subscribe or unsubscribe via the World Wide Web, visit http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies or, via email, send a message with subject or body 'help' to kernelnewbies-request(a)kernelnewbies.org You can reach the person managing the list at kernelnewbies-owner(a)kernelnewbies.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Kernelnewbies digest..." Message: 1 Date: Thu, 21 Jun 2012 18:17:36 +0800 From: ?? <lovelylich(a)gmail.com> Subject: Re: [RFC] tcp: How does SACK or FACK determine the time to start fast retransmition? To: Vijay Subramanian <subramanian.vijay(a)gmail.com> Cc: netdev <netdev(a)vger.kernel.org>, kernelnewbies(a)kernelnewbies.org Message-ID: <4FE2F4C0.4020900(a)gmail.com> Content-Type: text/plain; charset=UTF-8; format=flowed ? 2012/6/21 16:42, Vijay Subramanian ??: > With SACK, number of dupacks does not have much meaning. What matters is > --how the SACK scoreboard looks like i.e. which packets are tagged > Lost/Sacked/Retransmitted > -- Whether FACK is in use (this assumes holes in between sacked > packets are lost and have left the network and so we can send out more > packets) > > So, stack does not count the number of dupacks that have come in. Only > SACK blocks matter. > You can try to track the following path: > tcp_ack() deals with incoming acks and if it sees a dupack (does not > matter what number), or incoming packet contains SACK it calls > tcp_fastretrans_alert() which calls tcp_xmit_retransmit_queue(). > > tcp_xmit_retransmit_queue() decides which packets to retransmit. The > first packet to start retransmitting from is tracked in > tp->retransmit_skb_hint. > Note that the dupThresh is actually tracked by tp->reordering which > measures the reordering in the network and is not fixed at 3. So, if > more than > tp->reordering packets have been acked above a given packet, this > packet is a candidate for retransmisson. See tcp_mark_head_lost() to > see how the > reordering metric is used to mark packets as lost. This corresponds to > the check you mentioned in the RFC. > > So, window permitting, packets are sent as follows; > (a)-- Packets marked lost as per description above > (b)-- new packets (if any) > (c)-- Holes between sacked packets which are not reliably lost. > > choice between (b) and (c) is made in tcp_can_forward_retransmit(). > > Hope this helps. > Vijay > It is just I wanted! Thanks for your detailed explaination and kindness. ------------------------------ Message: 2 Date: Thu, 21 Jun 2012 17:57:22 +0530 From: Vijay Chauhan <kernel.vijay(a)gmail.com> Subject: Kernel Memory To: kernelnewbies(a)kernelnewbies.org Message-ID: <CAJ61zBA%2B%2BCFmTHSb2cRr%3DcKJdTkVYsSVs3_igESbbf21gWaqCQ(a)mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1 Hello, I am newbie. It has been said "kernel memory is not pageable" What does it mean? There is no concept of kernel virtual address? Any simple explanation will help me to udnerstand. Thanks, Vijay ------------------------------ Message: 3 Date: Thu, 21 Jun 2012 20:48:48 +0800 From: ?? <wangzhe5004(a)gmail.com> Subject: download the source code of some commands To: kernelnewbies <kernelnewbies(a)kernelnewbies.org> Message-ID: <CAJrRU3VHh719vZgCCJ3r8K842YemEo9d5_0swgqsGDNXE0vxVQ(a)mail.gmail.com> Content-Type: text/plain; charset="iso-8859-1" Hi all: I want to see some source code of some commands,for example,halt,reboot,uptime,and so on. but i don't kown where to download? can you give me some advice? Thanks in advance!

1 0

[RFC] tcp: How does SACK or FACK determine the time to start fast retransmition?
by 李易 June 21, 2012

June 21, 2012

HI all, When tcp uses reno as its congestion control algothim, it uses tp->sacked_out as dup-ack. When the third dup-ack(under default condition) comes, tcp will initiate its fast retransmition. But how about sack ? According to kernel source code comments, when sack or fack tcp option is enabled, there is no dup-ack counter. See comments for function tcp_dupack_heuristics(): http://lxr.linux.no/linux+v2.6.37/net/ipv4/tcp_input.c#L2300 So , how does tcp know the current dup-ack is the last one which triggers the fast retransmition? According to rfc3517 section 5: "Upon the receipt of the first (DupThresh - 1) duplicate ACKs, the scoreboard is to be updated as normal." "When a TCP sender receives the duplicate ACK corresponding to DupThresh ACKs, the scoreboard MUST be updated with the new SACK information (via Update ()). If no previous loss event has occurred on the connection or the cumulative acknowledgment point is beyond the last value of RecoveryPoint, a loss recovery phase SHOULD be initiated, per the fast retransmit algorithm outlined in [RFC2581]." But these sentences doesn't describe how tcp knows the current ack is the dup-threshold dup-ack. Accorrding to rfc3517 seciton 4 and isLost(Seqnum) function: "The routine returns true when either DupThresh discontiguous SACKed sequences have arrived above ’SeqNum’ or (DupThresh * SMSS) bytes with sequence numbers greater than ’SeqNum’ have been SACKed. Otherwise, the routine returns false." I think this is just what I am searching for, but I still don't know which line of code in Linux tcp protocol does this check. Can any one help me ? thks in advance.

2 2